Three essays on credit markets imperfections

Following the influential work of Stiglitz andWeiss (1981) a vast literature has recognized the prominent and pervasive role of information asymmetry in credit markets. In these markets information is distributed asymmetrically between agents: lenders may be unable to know ex ante either the creditworthiness of a potential borrower (adverse selection, (Stiglitz and Weiss, 1981)) or the degree of riskiness of a project (moral hazard, (Bester and Hellwig, 1987)); when lenders use the price mechanism to distinguish good from bad borrowers they face troublesome consequences: higher interest rates may induce ex-ante riskier individuals to join the pool of applicants or push borrowers to choose riskier investment projects. Thus lenders, instead of raising the credit cost, will rather choose to ration the applicants believed to be risky. This credit market imperfection constitutes a problem for firms that have to rely more on internal funds because raising external finance is either more costly or impossible. Because firms experiencing financing constraints cannot finance their desired level of investments as if internal and external funds were perfect substitutes, credit rationing may have large negative effects on firms’ investment, employment and asset growth (Ferrando and Mulier, 2015). In this dissertation, that consists of three self-contained research papers, I aim to study three different but related aspects of asymmetric information in explaining such credit market imperfections. Chapter 1 focuses on modelling and estimating the extent of credit rationing in a specific segment of the Italian credit market. In the chapter I analyze the role of credit demand and credit supply factors in unraveling the declining dynamics of bank loans to Italian manufacturing SMEs from 2010 to 2017. I look at potential imbalances between demand and supply of loans by means of a credit market disequilibrium model in which supply and demand are simultaneously estimated while the minimum of them is observed. I find that credit demand is positively associated to investment activity and negatively to cost of credit and to other external financing sources, while supply is higher for less risky and leveraged but more collateralized firms. Back-of-the-envelop calculations suggest that the declining trend in credit is mainly related to the drop in loans’ demand. Moreover, our estimated model allows us to compute a credit rationing measure for all firms in our sample and discuss the case for potential price rationing. I find that the calculated indices are in accordance with well established survey measures and are validated by empirical tests on the effects of financial constraints on investment dynamics. Asymmetric information between lenders and borrowers does not only imply potential rationing of quantities but may also lead to a suboptimal provision of long term credit by banks; this may have negative effects on firms’ investments and, as a consequence, future growth (Caprio Jr and Demirgüç-Kunt, 1998). In order to lessen the asymmetry lenders may ask borrowers to provide collateral to guarantee the loan. In Chapter 2 I analyze a policy intervention – Mutuo di Riassetto (MR) – launched by Trentino regional Government aimed at increasing firms’ debt maturity by means of a public guarantee scheme. Using a combination of difference–in–differences and instrumental variable approaches, I find that Mutuo di Riassetto had a temporary impact on debt maturity by raising firms’ share of long-term debt only for the first two years after the start of the program. The policy did not have relevant effects on performance: firms registered a short-run increase in intangible assets and (to a lesser extent) profitability, but did not display any permanent rise in terms of sales, tangible assets, labor cost, or credit access. I also find that firms involved in the MR program observed a relevant rise in the probability to default. In order to mitigate the effects of credit market failures, lenders use a number of screening technologies aimed at acquiring soft information that could substitute for hard data. In Chapter 3 I address the question if cultural proximity between lender and borrower possibly lessens informational problems that negatively affect lending. I use Credit Registry data from the population of loans granted to firms located in South Tyrol, a region hosting two different cultural groups to study the endogenous selection of firms and banks into same-culture lending transactions. I find that cultural proximity is a relevant trigger of lending relationships: Firms are more likely to demand credit from culturallyclose banks, and the first link that firms establish with the banking system is typically within the same cultural group. I find that information asymmetry, credit constraints, and credit market conditions explain an important share of the variation in borrowing from a same- vs. a different-culture bank. Overall, our findings suggest that banks and firms endogenously select into same- or different-culture relationships according to their needs to reduce information asymmetry or credit constraints.